Synthesis and QSAR study of novel α-methylene-γ-butyrolactone derivatives as antifungal agents

Article abstract of DOI:10.1016/j.bmcl.2017.01.030

Thirty-six new α-benzylidene-γ-lactone compounds based α-methylene-γ-butyrolactone substructure were prepared and characterized by spectroscopic analysis. All compounds were evaluated for antifungal activities in vitro against six plant pathogenic fungi and the half maximal inhibitory concentration (IC₅₀) against Botrytis cinerea and Colletotrichum lagenarium were investigated. Compounds 5c-3 and 5c-5 with the halogen atom exhibited excellent fungicidal activity against B. cinerea (IC₅₀?=?22.91, 18.89?μM). The structure-activity relationships (SARs) analysis indicated that the derivatives with electron-withdrawing substituents at the meta- or para-positions improves the activity. Via the heuristic method, the generated quantitative structure-activity relationship (QSAR) model (R²?=?0.961) revealed a strong correlation of antifungal activity against B. cinerea with molecular structures of these compounds. Meanwhile, the cytotoxicity of 20 representative derivatives was tested in the human tumor cells line (HepG2) and the hepatic L02 cells line, the result indicated that the synthesized compounds showed significant inhibitory activity and limited selectivity. Compound 5c-5 has the highest fungicidal activity with IC₅₀?=?18.89?μM (against B. cinerea.) but low cytotoxicity with IC₅₀?=?35.4?μM (against HepG2 cell line) and IC₅₀?=?68.8?μM (against Hepatic L02 cell line). These encouraging results can be providing an alternative, promising use of α-benzylidene-γ-lactone through the design and exploration of eco-friendly fungicides with low toxicity and high efficiency.

Full text of DOI:10.1016/j.bmcl.2017.01.030

Accepted Manuscript
Synthesis and QSAR study of novel α-methylene-γ-butyrolactone derivatives
as antifungal agents
Yong-Ling Wu, De-Long Wang, En-Hui Guo, Shuang Song, Jun-Tao Feng,
Xing Zhang
PII:
S0960-894X(17)30042-2
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.01.030
BMCL 24606
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
31 October 2016
7 January 2017
11 January 2017
Please cite this article as: Wu, Y-L., Wang, D-L., Guo, E-H., Song, S., Feng, J-T., Zhang, X., Synthesis and QSAR
study of novel α-methylene-γ-butyrolactone derivatives as antifungal agents, Bioorganic & Medicinal Chemistry
Letters (2017), doi: http://dx.doi.org/10.1016/j.bmcl.2017.01.030
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Title: Synthesis and QSAR study of novel α-methylene-γ-butyrolactone derivatives as
antifungal agents
Authorship:
Authors: Yong-Ling Wua, De-Long Wanga, En-Hui Guoa, Shuang Songa, Jun-Tao Feng a, b,*,
Xing Zhang a, b
a
Research and Development Center of Biorational Pesticide, Northwest A&F University,
Yangling 712100, Shaanxi , China.
b
Engineering and research center of biological pesticide of Shaanxi Province, Yangling
712100, Shaanxi , China.
*Corresponding Author: Telephone +86-29-87092122; fax +86-29-87092122; E-Mail:
fengjt@nwsuaf.edu.cn.
1

Abstract:
Thirty-six
new
α-benzylidene-γ-lactone
compounds
based
α-methylene-γ-butyrolactone substructure were prepared and characterized by
spectroscopic analysis. All compounds were evaluated for antifungal activities in vitro
against six plant pathogenic fungi and the half maximal inhibitory concentration (IC50)
against Botrytis cinerea and Colletotrichum lagenarium were investigated. Compounds
5c-3 and 5c-5 with the halogen atom exhibited excellent fungicidal activity against B.
cinerea (IC50 = 22.91, 18.89 μM). The structure-activity relationships (SARs) analysis
indicated that the derivatives with electron-withdrawing substituents at the meta- or
para-positions improves the activity. Via the heuristic method, the generated quantitative
structure-activity relationship (QSAR) model (R2 = 0.961) revealed a strong correlation of
antifungal activity against B. cinerea with molecular structures of these compounds.
Meanwhile, the cytotoxicity of 20 representative derivatives was tested in the human tumor
cells line (HepG2) and the hepatic L02 cells line, the result indicated that the synthesized
compounds showed significant inhibitory activity and limited selectivity. Compound 5c-5
has the highest fungicidal activity with IC50 = 18.89 µM (against B. cinerea.) but low
cytotoxicity with IC50 = 35.4 µM (against HepG2 cell line) and IC50 = 68.8 µM (against
Hepatic L02 cell line). These encouraging results can be providing an alternative,
promising use of α-benzylidene-γ-lactone through the design and exploration of
eco-friendly fungicides with low toxicity and high efficiency.
Keywords: α-benzylidene-γ-lactone derivatives; antifungal activity; QSAR; heuristic
method; cytotoxicity
2

Plant fungal pathogens have significant impacts on the crops, and often lead to
significant yield reduction and dramatic economic losses in agriculture.1-2 Furthermore,
many of the fungi can produce mycotoxins harmful to animal and human health.3
Meanwhile, many commercial chemical fungicides have several detrimental effects, such as
residual toxicity, severe pesticide resistance, and environmental pollution.4 Thus, there is a
growing need to develop new antifungal agents to effectively control agricultural diseases.
Using natural products as lead compounds to develop new pesticides with novel
structures and mechanism is one of the most effective methods for pesticide design.
Sesquiterpene lactones (STLs) have been considered interesting leads to develop a new
class of potential agents in the past, such as costunolide, andrographolide and carabrone
(isolated from fruits of Carpesium macrocephalum) (Figure 1).5-7 The functional unit of
α-methylene-γ-butyrolactone substructure is one of the commonly chemical scaffolds
among numerous natural products because it’s electrophilic α, β-unsaturated carbonyl
structure could react with biological nucleophiles.8-9 STLs natural products containing the
α-methylene-γ-butyrolactone substructure always were found to possess a broad spectrum
of biological activities, including anticancer, antibacterial, anti-inflammatory, and
antimicrobial properties.10-13 While, great biological activities of this class of compounds
have been extensively reported in the pharmaceutical field but not in the agrochemical
field.14 In our previous research, we demonstrated carabrone and a series of carabrol
derivatives exhibited prominent antifungal activity, and we also found aromatic substituents
directly fused to the γ-position of the α-methylene-γ-lactone ring boost their antifungal
potency more effectively.15-16 Therefore, compounds containing α-methylene-γ-lactone can
3

be used to develop novel and improved crop-protection agents.
O
O
O
O
O
O
OH
OH
O
carabrone
costunolide
andrographolide
Figure 1. Chemicial structures of the representative sesquiterpene lactones
While, during the design and modification of potential antifungal agents derived from
α-methylene-γ-butyrolactone chemical scaffolds, the exocyclic carbon-carbon double bond
in α, β-unsaturated carbonyl system need to be considered. In order to analysis the roles of
electron density and steric hindrance influence on the exocyclic carbon-carbon double bond,
a series of α-benzylidene-γ-lactone were prepared and their activity were evaluated in this
study. In addition, developing and screening candidates with antifungal activity from
thousands of compounds is virtually and economically impossible, and employing
quantitative structure-activity relationship (QSAR) study is a method to reduce the
problems on cost and time requirements for screening.17 Meanwhile, after the essential
structural features for the activity were defined, the mechanism of action can be elaborated
advantageously. 18-19
In continuation of our investigation on the design and synthesis of bioactive
compounds, six important crops threatening pathogenic fungus in agriculture were choosed
to initial screen all the compounds antifungal activities, and with the higher preliminary
activity the half maximal inhibitory concentration (IC50) against B. cinerea and C.
4

lagenarium were investigated. Meanwhile, the cytotoxicity was tested to ensure selectivity
of the antifungal effects. In addition, a QSAR study was also performed on all of the
derivatives using some quantification software packages, which can correlate their
structural features with their antifungal activities.
OH
1
R
a
COOH
COOH
R1CHO
Br
H
1
2
3
b
O
O
O
O
c
1
H
R
1
R
2
R
5
4
5： R1= ph, 4-phCH3,4-phCl,
R2= ph, 2-phBr, 3-phCl, 2-phF,
3-phF, 3-phCH3O, 4-ph(CH3)3,
3-phCH3, Naphth, 2-ph(CF3)3 ,
4-phCH3CH2O or 2-phNH2
4a: R1= ph
4b: R1= 4-phCH3
4c: R1= 4-phCl
scheme 1. Synthesis route of the target compounds 5.
Reagents and conditions: (a) indium powder, THF, rt. (b) 6 M HCl, rt, 6 h. (c) iodobenzene, Pd(OAc)2,
Et3N, DMF, 80°C, rt.
As shown in scheme 1, three kinds of intermediate compounds 4a-4c were obtained
under milder aqueous reaction conditions through indium-mediated Barbier allyl addition
to aldehydes.20 This versatile and simple method was used to prepare γ-substituted
α-methylene-γ-butyrolactones in good yields (72–93%) under markedly milder aqueous
reaction conditions than other reported methods.15 The palladium-catalyzed arylation of
different α-methylene-γ-lactone lactones was shown to produce E-olefin coupling products
selectively in moderate to excellent yields.21 Herein, through the previously reported Heck
5

reaction conditions (0.04 mmol of Pd(OAc)2 with Et3N in DMF at 80 °C), substituted
α-methylene-γ-butyrolactones (4a-4c) were reacted with a series of readily available aryl
iodides to obtained corresponding α-benzylidene-γ-lactones compounds 5 (table 1). We also
strategically selected aryl iodides that contained different substitution pattern and the
presence of electron-donating or electron-withdrawing substituents on the aromatic ring to
understand the structure-activity relationships of substituents on the aromatic ring of the
analogues. Meanwhile, the substituents on the aromatic ring did not affect the reaction
1
13
yields. Moreover, the structures of all the derivatives were characterized by H NMR, C
NMR and high-resolution electrospray ionization mass spectrometry (HR-ESI-MS).
Table 1 Initial Antifungal Activity of Compounds 5 at100 μg/mL a
Compound
average inhibition rate (%) (100 μg/mL; 72h)
no.
5a-1
R1
H
H
H
H
H
H
H
H
H
H
R2
ph
B. c.
C. l.
P. c.
F. o.
S. s.
F. g.
1
2
3
4
5
6
7
8
9
80.1 ± 0.7
81.0 ± 0.1
88.9 ± 0.2
86.8 ± 0.3
91.7 ± 0.1
75.2 ± 0.6
71.3 ± 0.2
76.8 ± 0.2
83.2 ± 0.8
83.9 ± 0.9
67.7 ± 0.2
69.1 ± 0.3
75.0 ± 0.5
83.8 ± 0.5
85.7 ± 0.1
85.1 ± 0.3
89.3 ± 0.2
70.5 ± 0.8
67.0 ± 0.5
74.2 ± 0.2
76.2 ± 0.3
77.3 ± 0.5
85.5 ± 0.3
81.6 ± 0.8
86.2 ± 0.5
67.9 ± 0.3
65.1 ± 0.7
71.7 ± 0.1
68.7 ± 0.2
77.2 ± 0.4
65.9 ± 0.3
66.6 ± 0.8
75.2 ± 0.7
81.8 ± 0.3
84.2 ± 0.5
83.3 ± 0.9
85.6 ± 0.5
74.4 ± 0.3
71.5 ± 0.4
75.7 ± 0.7
6
48.3 ± 0.8
46.9 ± 0.4
49.5 ± 0.5
51.7 ± 0.7
65.8 ± 0.9
46.3 ± 0.0
44.6 ± 0.2
49.5 ± 0.1
53.8 ± 0.4
53.2 ± 0.3
46.6 ± 0.6
43.5 ± 0.8
48.7 ± 0.6
54.2 ± 0.4
56.3 ± 0.1
55.1 ± 0.3
60.6 ± 0.6
46.2 ± 0.8
44.2 ± 0.5
49.6 ± 0.6
53.3 ± 0.3
51.7 ± 0.3
54.8 ± 0.6
57.3 ± 0.0
64.6 ± 0.9
47.3 ± 0.8
46.9 ± 0.6
54.8 ± 0.2
53.5 ± 0.4
56.6 ± 0.6
47.8 ± 0.9
45.2 ± 0.7
50.5 ± 0.9
53.6 ± 0.4
55.8 ± 0.6
55.6 ± 0.1
58.1 ± 0.2
47.3 ± 0.2
43.3 ± 0.3
49.6 ± 0.5
70.3 ± 0.1
72.2 ± 0.3
73.8 ± 0.4
71.9 ± 0.4
77.3 ± 0.6
65.5 ± 0.4
64.6 ± 0.3
66.1 ± 0.1
65.0 ± 0.2
73.8 ± 0.8
58.7 ± 0.3
62.2 ± 0.5
67.1 ± 0.6
70.3 ± 0.9
73.2 ± 0.2
72.7 ± 0.1
75.3 ± 0.1
64.6 ± 0.4
61.1 ± 0.5
65.3 ± 0.8
67.7 ± 0.7
68.3 ± 0.5
74.5 ± 0.6
70.6 ± 0.5
79.2 ± 0.3
64.3 ± 0.6
60.9 ± 0.4
67.1 ± 0.2
71.8 ± 0.1
69.7 ± 0.0
62.4 ± 0.8
61.6 ± 0.7
65.1 ± 0.4
66.5 ± 0.5
73.7 ± 0.2
70.2 ± 0.4
76.2 ± 0.8
57.7 ± 0.4
55.6 ± 0.6
63.7 ± 0.7
5a-2
5a-3
5a-4
5a-5
5a-6
5a-7
5a-8
5a-9
2-phBr
3-phCl
2-phF
3-phF
3-phCH3O
4-phC(CH3)3
3-phCH3
2-naphthyl
2-phCF3
10 5a-10
11 5a-11
12 5a-12
H
4-phCH3CH2O
2-phNH2
ph
H
13
14
15
16
17
18
19
20
5b-1
5b-2
5b-3
5b-4
5b-5
5b-6
5b-7
5b-8
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
2-phBr
3-phCl
2-phF
3-phF
3-phCH3O
4-phC(CH3)3
3-phCH3

21
5b-9
CH3
CH3
2-naphthyl
2-phCF3
77.5 ± 0.9
81.3 ± 0.4
65.1 ± 0.6
66.2 ± 0.7
80.7 ± 0.8
87.3 ± 0.4
90.9 ± 0.1
89.2 ± 0.3
97.6 ± 0.4
79.5 ± 0.7
77.3 ± 0.9
81.2 ± 0.4
82.8 ± 0.5
85.7 ± 0.3
76.6 ± 0.8
77.3 ± 0.2
19.4 ± 0.9
68.1 ± 0.6
84.0 ± 0.3
65.6 ± 0.4
65.4 ± 0.0
77.3 ± 0.2
84.6 ± 0.4
87.5 ± 0.6
87.1 ± 0.8
93.0 ± 0.1
75.3 ± 0.2
76.2 ± 0.4
77.4 ± 0.5
77.8 ± 0.7
80.6 ± 0.6
75.1 ± 0.3
76.5 ± 0.5
21.5 ± 0.8
45.2 ± 0.3
49.6 ± 0.7
43.2 ± 0.4
46.3 ± 0.2
50.6 ± 0.1
57.7 ± 0.6
63.8 ± 0.6
60.6 ± 0.3
70.7 ± 0.7
50.5 ± 0.4
49.3 ± 0.5
52.2 ± 0.2
50.6 ± 0.1
55.7 ± 0.2
48.2 ± 0.6
49.3 ± 0.3
25.1 ± 0.8
47.6 ± 0.6
51.6 ± 0.8
41.2 ± 0.2
47.9 ± 0.6
51.2 ± 0.4
57.0 ± 0.4
60.1 ± 0.8
59.4 ± 0.7
68.9 ± 0.4
45.8 ± 0.7
44.7 ± 0.0
47.3 ± 0.3
55.8 ± 0.4
54.6 ± 0.8
43.6 ± 0.7
45.6 ± 0.2
17.8 ± 0.9
66.4 ± 0.5
68.8 ± 0.7
47.2 ± 0.3
59.6 ± 0.2
63.7 ± 0.1
73.8 ± 0.6
76.1 ± 0.5
74.0 ± 0.3
77.3 ± 0.2
67.6 ± 0.6
52.5 ± 0.1
68.2 ± 0.6
72.7 ± 0.8
67.6 ± 0.0
52.5 ± 0.2
53.4 ± 0.4
13.3 ± 0.6
64.2 ± 0.2
67.1 ± 0.3
47.7 ± 0.3
54.2 ± 0.5
67.3 ± 0.8
71.4 ± 0.4
77.8 ± 0.6
75.1 ± 0.3
83.8 ± 0.8
65.1 ± 0.5
54.0 ± 0.2
70.2 ± 0.3
70.7 ± 0.6
72.4 ± 0.7
53.3 ± 0.3
55.8 ± 0.1
22.8 ± 0.3
22 5b-10
23 5b-11
24 5b-12
CH3 4-phCH3CH2O
CH3
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
2-phNH2
ph
25
26
27
28
29
30
31
32
33
5c-1
5c-2
5c-3
5c-4
5c-5
5c-6
5c-7
5c-8
5c-9
2-phBr
3-phCl
2-phF
3-phF
3-phCH3O
4-phC(CH3)3
3-phCH3
2-naphthyl
2-phCF3
4-phCH3CH2O
2-phNH2
34 5c-10
35 5c-11
36 5c-12
37
Tulipalin Ab
Note: a B. c., Botrytis cinerea; C.l., Colletotrichum lagenarium; P.c. ,Phytophthora capsici; F.o., Fusarium
oxysporum.sp.cucumebrium; S.s., Sclerotinia sclerotiorum; F.g., Fusarium graminearum. b Natural lead compound
tulipalin A was used as the positive control.
According to the mycelium linear growth rate method reported previously, all
α-benzylidene-γ-lactones compounds 5 were screened for antifungal activity against six
plant pathogenic fungi in vitro at 100 μg/mL. The results are listed in Table 1. Almost all of
the test compounds exhibited some inhibition activity against each of the fungi at 100
μg/mL, and all derivatives exhibited higher activity than the natural lead compound
tulipalin A. Among these compounds, compounds 5a-5, 5b-3-5, 5c-2-5 showed broad
spectra antifungal activities against all tested phytopathogens, especially, most compounds
were more active against B. cinerea and C. lagenarium. On the basis of these results, all
compounds with the higher preliminary activity against B. cinerea and C.lagenarium were
further assayed for the half maximal inhibitory concentration (IC50).
The results of the antifungal activity against B. cinerea are summarized in Table 2,
7

from which it can be seen that the substitution pattern and the halogen atom containing
derivatives exhibited significant antifungal activity against B. cinerea. The following four
main structure-activity relationships (SARs) obtained. First, it is easy to see that all the
compounds (5a-1-12, 5b-1-12, and 5c-1-12) were found to have higher activity than the
corresponding intermediate compounds (4a, 4b, and 4c). Among all the derivatives,
compounds 5c with a chlorine atom intermediate were more active than those of the other
compounds (5a and 5b, respectively). Second, the introduction of the electron-withdrawing
groups F, Cl, and Br to the benzene ring dramatically increased the potency. Compounds
5a-2-5, 5b-2-5, and 5c-2-5 (IC50 = 18.89-44.67 μM) exhibited antifungal activity
approximately four to ten-fold higher than intermediate compounds (4a, 4b and 4c,
respectively). It was notable that the IC50 values of 5c-3 and 5c-5 were approximately
threefold less than carbendazim. While, the electron-donating groups CH3, C(CH3)3, NH2,
CH3O and CH3CH2O introduced to the benzene ring to give 5a-6-8, 5a-11-12, 5b-6-8,
5b-11-12, 5c-6-8, and 5c-11-12, (IC50 = 42.69-125.89 μM) greatly weakened the potency.
Third, the result suggested that the steric hindrance on the benzene ring have an important
influence on the antifungal activity. It is worth noting that the compounds 5c-10 (-CF3)
(IC50 = 33.88 μM) containing an ortho substituent on the aryl ring was found less active
than compounds 5c-5 (-F) (IC50 = 18.89 μM) with a meta substituent. Also, the presence of
a t-Bu group (5a-7, 5b-7 and 5c-7) on the benzene ring decreases the activity in all cases.
Fourth, the effects of aromatic properties of the compounds have an important influence on
the antifungal activity. Compound (5a-9, 5b-9 and 5c-9) with the alphy-naphthyl group
increases the activity in all or most cases. Overall, the derivatives with
8

electron-withdrawing substituents at the meta- or para-positions improves the activity, in
contrast, the presence of an electron-donating group at the ortho-positions on the aryl ring
decreases the activity.
Table 2 In vitro fungicidal activity of compounds against B.cinerea and C. lagenarium.
B. cinerea
C. lagenarium
B. cinerea
C. lagenarium
Compd.
Compd.
a
a
a
a
IC50
,
IC50
,
IC50
μM
,
IC50
,
pIC50
pIC50
μM
μM
μM
-1.79
-1.58
-1.44
-1.54
-1.42
-1.85
-1.92
-1.82
-1.71
-1.59
-1.99
-1.93
-1.86
-1.65
-1.49
-1.62
-1.46
-1.98
-1.99
-1.89
74.37
46.95
35.56
46.59
36.08
77.90
93.42
75.16
59.98
47.25
106.08
96.94
81.84
52.30
38.67
55.10
42.50
103.19
110.88
87.01
-1.79
-1.74
-2.10
-2.01
-1.55
-1.47
-1.36
-1.44
-1.22
-1.65
-1.69
-1.62
-1.57
-1.53
-1.77
-1.75
-2.22
-2.29
-2.02
-0.92
73.99
61.00
137.61
112.18
43.54
38.32
27.99
38.09
29.18
64.00
53.21
51.08
47.80
45.96
67.36
61.88
189.00
215.33
128.83
10.99
5b-9
5b-10
5b-11
5b-12
5c-1
61.66
38.02
27.54
34.67
26.30
63.10
83.18
66.07
51.29
38.90
97.72
85.11
72.44
44.67
30.90
41.69
28.84
95.50
97.72
77.62
61.66
54.95
125.89
102.33
35.48
29.51
22.91
27.54
18.89
44.67
52.98
42.69
37.15
33.88
58.88
56.23
164.30
196.19
104.31
8.38
5a-1
5a-2
5a-3
5a-4
5a-5
5a-6
5a-7
5a-8
5a-9
5a-10
5a-11
5a-12
5b-1
5b-2
5b-3
5b-4
5b-5
5b-6
5b-7
5b-8
5c-2
5c-3
5c-4
5c-5
5c-6
5c-7
5c-8
5c-9
5c-10
5c-11
5c-12
4a
4b
4c
b
carbendazim
Note: a All half maximal inhibitory concentration (IC50) values are presented as the means ± SD (n = 3),
µM; b Commercial fungicide, carbendazim was used as the positive control.
The results of the antifungal activity against C. lagenarium are summarized in Table 2,
from which we can see that compounds 5a-2-5, 5b-3, 5b-5, 5c-1-5, and 5c-9-10 exhibited
moderate antifungal activity against C. lagenarium. Most of the test compounds showed
less effective than against B. cinerea.
9

Figure 2 The “breaking point” rule results
After conformer optimizing, minimum energy calculating and format converting, the
best regression relation between antifungal activity against B. cinerea and descriptors can
be established. Therefore, it is necessary for selecting exact regression analysis method in
CODESSA 2.7.15 software, which has a large number of regression analysis methods like
heuristic regression analysis, multi-linear regression analysis, etc. In this paper, 36
α-methylene-γ-butyrolactones analogues were used as samples and 5 groups of descriptors
were calculated. In view of the above results, the heuristic regression analysis method was
considered for constructing the QSAR model. The number of the descriptors was confirmed
using the “breaking point” graph rule (Figure. 2). The heuristic regression analysis
indicated a statistically significant improvement in the correlation coefficient (R2 = 0.986)
when the descriptors varied from 2 to 5. Nevertheless, no obvious improvement (R2 = 0.977)
was observed when the descriptors varied from 5 to 10. After the number of the descriptors
reached a certain value, this statistical improvement in regression equation became less
unimportant (△R2 < 0.02–0.04).22 Meanwhile, the numbers of samples and descriptors also
meet the equation (3D ≤ S-3, S means the number of samples; D means the number of
10

descriptors). Thus, the final 5-descriptor model was generated according to the above
results.
Table 3 The difference between the experimental pIC50 and predicted pIC50
No.
1
Compd.
5a-1
5a-2
5a-3
5a-4
5a-5
5a-6
5a-7
5a-8
5a-9
5a-10
5a-11
5a-12
5b-1
5b-2
5b-3
5b-4
5b-5
5b-6
5b-7
5b-8
5b-9
5b-10
5b-11
5b-12
5c-1
Calc.pIC50
-1.7563
-1.5405
-1.4717
-1.5907
-1.4460
-1.8556
-1.9483
-1.8206
-1.7431
-1.5514
-1.9558
-1.9006
-1.8145
-1.6291
-1.4476
-1.6779
-1.4744
-1.9897
-1.9657
-1.8666
-1.7947
-1.7663
-2.0828
-2.0473
-1.4967
-1.4395
-1.3633
-1.5062
-1.3513
-1.6753
-1.6972
-1.5774
-1.5059
-1.4896
-1.7927
-1.7476
Exp.pIC50
-1.7900
-1.5800
-1.4400
-1.5400
-1.4200
-1.8500
-1.9200
-1.8200
-1.7100
-1.5900
-1.9900
-1.9300
-1.8600
-1.6500
-1.4900
-1.6200
-1.4600
-1.9800
-1.9900
-1.8900
-1.7900
-1.7400
-2.1000
-2.0100
-1.5500
-1.4700
-1.3600
-1.4400
-1.3200
-1.6500
-1.6900
-1.6200
-1.5700
-1.5300
-1.7700
-1.7500
Difference
0.0337
0.0395
-0.0317
-0.0507
-0.0260
-0.0056
-0.0283
-0.0006
-0.0331
0.0386
0.0342
0.0294
0.0455
0.0209
0.0424
-0.0579
-0.0144
-0.0097
0.0243
0.0234
-0.0047
-0.0263
0.0172
-0.0373
0.0533
0.0305
-0.0033
-0.0662
-0.0313
-0.0253
-0.0072
0.0426
0.0641
0.0404
-0.0227
0.0024
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
5c-2
5c-3
5c-4
5c-5
5c-6
5c-7
5c-8
5c-9
5c-10
5c-11
5c-12
11

This optimized model showed that the predicted values of pIC50 (negative log IC50)
can be calculated, which was shown in Table 3. Moreover, the comparison chart of
predictive and practical activity of 36 derivatives is shown in Figure. 3. This model
included five descriptors in descending order according to their statistical significance (t
values), which is shown in Table 4, and the regression coefficients X and their standard
errors ∆X are also listed. The final QSAR model with 5 descriptors was shown in Eq. (1) as
pIC50 = -8.6656 + 2.7387 × MAOEP - 0.1016 ×µ + 0.6023 ×qC + 0.8535 ×no -
min
2.3901× qO
(1)
max
N= 36, R2 = 0.961, F = 84.05, S2 = 0.0021
Figure 3. Experimental pIC50 vs. predicted pIC50.
Table 4 The best five-descriptor model
Descriptor No.
X
±△X
t-Text
13.8371
-7.6452
8.8830
-1.0588
-4.5835
-2.6836
Descriptor
Intercept
0
1
2
3
4
5
-8.6656
6.2626×10-1
3.5882×10-1
1.1442×10-2
5.6884×10-1
1.8621×10-1
2.7387
MAOEP a
-1.0164×10-1
6.0228×10-1
8.5350×10-1
-2.3901
µ b
c
qC
min
d
no
8.9065×10-1
qO
e
max
b
c
Note: a Max. atomic orbital electronic population. The total dipole moment of the molecule. Min. net
d
e
atomic charge for a C atom. Number of occupied electronic levels of atoms. Max. net atomic charge
for a O atom.
12

The internal validation and the “leave-more-out” cross-validation methods were used
to validate the developed QSAR model.23 The internal validation was carried out by
dividing the compound data into three subsets A–C. The compounds 1, 4, 7, 10, etc., went
into the first subset (A); 2, 5, 8, 11, etc., went into the second subset (B); and 3, 6, 9, 12,
etc., went into the third subset (C). Two of the three subsets, (A and B), (A and C), and (B
and C), consist the training set while the remaining subset was treated as a test set. The
correlation equations were derived from each of the training sets using the same descriptors
and then used to predict values for the corresponding test set.24 Internal validation results
2
2
are presented in Table 5. The RTraining and RTest are within 5% for all three sets, and the
2
2
average values of RTraining = 0.964 and RTest = 0.963 were close to the overall R2 value.
Thus, the obtained QSAR model obtained demonstrated the predictive power of 3-fold
cross-validation. Meanwhile, the “leave-more-out” method was completed in a similar
manner to the internal validation. Every fourth compound (1, 5, 9, 13, etc.) was put into an
external test set, and the remaining compounds were left in the training set. The QSAR
model containing the same five descriptors was obtained with R2 = 0.968 from the training
set. When the same QSAR model was applied on the test set, R2 = 0.959 was observed.
Hence, the “leave-more-out” cross-validation results were also satisfactory.
Table 5 Internal validation of the QSAR model a
Training set
A+B
N
R2
F
S2
Test set
N
R2
F
S2
24
24
24
0.958
0.969
0.965
0.964
91.56
87.87
82.83
87.42
0.0028
0.0022
0.0024
0.0025
C
A
12
12
12
0.967
0.958
0.963
0.963
84.65
83.82
87.31
85.26
0.0026
0.0024
0.0025
0.0025
B+C
A+C
B
Average
Average
Note: a Compds. A: 1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31, 34. Compds. B: 2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 32, 35.
Compds. C: 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 33, 36.
13

Some insights into the structure of α-methylene-γ-butyrolactones derivatives structure
that influences antifungal activity may be gained by explaining the descriptors referred to
the constructed QSAR model. The 1st and 4th most important descriptors obtained in the
model were the maximum atomic orbital electronic population and the number of occupied
electronic levels of atoms, which have a significant effect on the antifungal activity.
Maximum atomic orbital electronic population for a given atomic species in the molecule is
an important index to describe the nucleophilicity of the molecule, which is directly related
to molecular nucleophilic capacity and characterizes the susceptibility of the molecule to
electrophilic attack.25 The number of occupied electronic levels of atoms depends directly
on the quantum-chemically calculated charge distribution in the molecules, and therefore
describes
the
polar
interactions
between
molecules.26-27
In
fact,
α-methylene-γ-butyrolactones derivatives with the α, β-unsaturated carbonyl system
(Michael acceptor), which had higher electron deficiency induced by electron-withdrawing
groups, can be easily attacked by bionucleophiles.12-13 In eqn (1), these two descriptors
appearance with a positive sign in this model, and the electron withdrawing substitution
groups in a molecule with a higher descriptor value would has a higher pIC50, which
indicated the obtained QSAR study result partially met the above SAR study conclusion.
The second important descriptor was the total dipole moment of the molecule. This
descriptor was important in modulating antifungal activity because of the presence of C=O
in the molecule, which exhibited permanent polarization due to an electronegativity
difference between the atoms.28-29 The C(C=O) atoms may be involved in binding
interactions with cells present at the target site. The total dipole moment of the molecule
14

thus played a critical role in modulating the antifungal of the test compounds. In Eq. (1),
appearance with a positive sign in the model indicated that molecule with higher descriptor
value had a higher pIC50. In contrary, a negative sign in the model indicated that molecule
with lower descriptor value had a higher pIC50.
Figure 4 Optimal conformer, charge distribution and contour map of compounds 5a-5and 5a-8
The 3rd and 5th descriptors obtained in the model were min. net atomic charge for a C
atom and max. net atomic charge for a O atom. These two descriptors belonged to
electrostatic descriptors and they reflect the charge distribution of the molecules as shown
in the contour maps (Figure. 4), the green lines represent electron density decrease part, and
the red lines represent electron density increase part which exhibits the strongest electron
attraction.30-31 Meanwhile, as the presence of the of electron activity difference between the
atoms, the permanent polarization was shown in the molecular electrostatic potential map
(Figure. 5), the exocyclic carbon-carbon double bond exhibit the greater negative
electrostatic potential which was easily occurred necleophilic reaction.32-33 This finding was
consistent with those of electrophilic α, β-unsaturated carbonyl structure in the
α-methylene-γ-butyrolactone could reaction with biological nucleophiles.12-13 These
observations suggested that the electrostatic properties of C atom and O atom were
15

important elements affect the antifungal activity of the test compounds.
Figure 5 molecular electrostatic potential map of compounds 5a-3 and 5a-5
5a-3
5a-5
In general, sesquiterpene lactone with the α-methylene- γ-butyrolactone structure often
showed a high toxicity potential against mammalian cells.34-35 In order to ensure the
selectivity of the antifungal effects, the cytotoxicity of 20 representative derivatives was
tested in the human tumor cells line (HepG2) and Hepatic L02 cells. The result is listed in
Table 6, which indicated that the QSAR underlying the antifungal and cytotoxic effects of
these representative compounds are different. For instance, compound 5c-5 has the highest
fungicidal activity with IC50 = 18.89 µM (against B. cinerea.) but low activity with IC50 =
35.4 µM (against HepG2 cell line) and IC50 = 68.8 µM (against Hepatic L02 cell line). On
the contrary, low antifungal activity compound 5a-6 (IC50 = 83.18 µM) has moderate
cytotoxic activity with IC50 = 18.7µM (against HepG2 cell line) and IC50 = 42.1 µM
(against Hepatic L02 cell line). The results also showed that compound 5c-5 displayed high
selectivity for the activity against Hepatic L02 cell line to the B. cinerea. (selectivity
index>3).
Through QSAR studies on antifungal and
cytotoxicity of
16

α-benzylidene-γ-butyrolactone derivatives, these are important points that need further
investigation to seek high activity derivatives with non-cytotoxicity.
Table 6 In vitro fungicidal activity of compounds against B.cinerea and cytotoxic activity against
human tumor cells lines (HepG2) and Hepatic L02 cells
IC50 ( μM )
(against
HepG2 cell
line)
IC50 ( μM )
(against
Selectivity index
IC50 ( μM )
(against
(activity against
No.
Compd.
Hepatic L02
Cells )
Hepatic L02 Cells
B.cinerea )
/ B.cinerea)
1
2
5a-1
5a-3
5a-4
5a-6
5a-7
5a-10
5b-2
5b-4
5b-5
5b-8
5b-9
5b-10
5c-3
5c-5
5c-7
5c-9
5c-11
5c-12
4a
61.66
27.54
34.67
83.18
66.07
38.90
44.67
41.69
28.84
77.62
61.66
54.95
22.91
18.89
52.98
37.15
58.88
56.23
164.30
104.31
32.6
44.8
35.3
18.7
38.3
45.3
24.5
33.1
21.7
42.2
85.2
98.6
43.6
35.4
78.4
55.2
20.9
>106.2
>128.6
87.9
44.5
25.3
56.6
42.1
24.0
26.9
63.6
45.6
32.8
77.6
>112.6
78.5
55.9
68.8
89.9
27.1
65.6
96.2
>166.1
84.8
<1
<1
1
3
4
<1
5
<1
<1
1
6
7
8
1
9
1
10
11
12
13
14
15
16
17
18
19
20
<1
1
1
2
3
1
<1
1
1
1
4c
<1
In this study, 36 α-benzylidene-γ-butyrolactone derivatives, namely, 5a-1-12, 5b-1-12
and 5c-1-12 were synthesized. The antifungal activity results showed compounds 5c with a
chlorine atom intermediate were more active than those of the other compounds.
Meanwhile, the antifungal activity against B. cinerea and C.lagenarium of compounds 5c-3
17

and 5c-5 was excellent. Moreover, the result of the SARs and QSAR studies exhibited that
the higher electron density around the α-methylene-γ-butyrolactone backbone structure and
smaller steric hindrance on the benzene ring played great beneficial effect on the antifungal
activity. Although all the compounds were not so effective against B. cinereal and
C.lagenarium compared with the positive control, the promising results obtained from
SARs, QSAR and cytotoxicity studies based α-benzylidene-γ-butyrolactone derivatives will
inspire us to carry on further work for seek the high-activity and low-toxic fungicides
candidate.
Acknowledgments
We greatly appreciate the funding support for this research provided by the National
Natural Science Foundation of China (No. 31471800). We also thank Kunming Institute of
Botany, Chinese Academy of Science for the HR-MS spectral data.
Supporting Information
Supplementary data associated with this article can be found, in the online version,
at …
Reference and notes
1. Strange, R. N.; Scott, P. R. Annu. Rev. Phytopathol. 2005, 43, 83-116.
2. Wilson, R. A.; Talbot, N. J. Microbiology (London, U. K.) 2009, 155, 3810 −3815.
3. Brase, S.; Encinas, A.; Keck, J.; Nising, C. F. Chem. Rev. 2009, 109, 3903−3990.
4. Dodds, P.N.; Rathjen, J.P. Nat. Rev. Genet. 2010, 11, 539–548.
5. Neerman, M. F. Int. J. Aromather. 2003, 13, 114-120.
6. Feng, J.T.; Wang, H.; Ren, S.X.; He, J.; Liu, Y.; Zhang, X. J. Agric. Food Chem. 2012, 60,
3817–3823.
7. Feng, J.T.; Zhang, Y.M.; Wang, J.R.; Zhang, X. Chin. J. Pestic. Sci. 2007, 9, 185–188.
18

8. Antonio, G.G.; Margarita, H.S.; Juan, I.P. J. Med. Chem. 2002, 45, 2358–2361.
9. Miyazawa, M.; Shimabayashi, H.; Hayashi, S.; Hashimoto, S.; Nakamura, S.; Kosaka, H.; Kameoka,
H. J. Agric. Food Chem. 2000, 48, 5406–5410.
10. Han, C.; Barrios, F.J.; Mark, V.R.; David, A.C. J. Org. Chem. 2009, 74, 7176–7179.
11. Irakusne, L.; Santiago, R.; Javier, I.; Florenci, V. J. Org. Chem. 2007, 72, 6614–6617.
12. Ellis, G.W.L.; Tavares, D.F.; Rauk, A. Can. J. Chem. 1985, 63, 3510–3515.
13. Lavallee, J.F.; Berthiaume, G.; Deslongchamps, P. Tetrahedron Lett. 1986, 27, 5455–5458.
14. Schmidt, T. J. Curr. Org. Chem. 1999, 3, 577.
15. Feng, J.T.; Wang, D.L.; Wu, Y.L.; Yan, H.; Zhang, X. Bioorg. Med. Chem. Lett. 2013, 23,
4393–4397.
16. Wu, Y.L.; Wang, D.L.; Gao, Y.Q.; Feng, J.T.; Zhang, X. Molecules. 2016. 21, 130.
17. Gao, Y.; Li, J.; Song, Z.; Song, J.; Shang, S.; Xiao, G.; Wan, Z.; Rao, X. Ind. Crop. Prod. 2015, 76,
660–671.
18. García, I.; Fall, Y.; Gómez, G.; González-Díaz, H. Mol. Divers. 2011, 15, 561–567.
19. Prado-Prado, F.J.; Borges, F.; Perez-Montoto, L.G.; González-Díaz, H. Eur. J. Med. Chem. 2009, 44,
4051–4056.
20. Choudhury, P.K.; Foubelo, F.; Yus, M. Tetrahedron. 1999, 55, 10779–10788.
21. Changho Han;Francis J. Barrios; Mark V. Riofski; David A. Colby. J. Org. Chem. 2009, 74,
7176–7179.
22. Wang, Z.; Song, J.; Han, Z.; Jiang, Z.; Zheng, W.; Chen, J.; Song, Z.; Shang, S. J. Agric. Food Chem.
2008.56, 11361–11366.
19

23. Liao, S.; Song, J.; Wang, Z.; Chen, J.; Fan, G.; Song, Z.; Shang, S.; Chen, S.; Wang, P. Bioorg. Med.
Chem. Lett. 2014, 24, 773–779.
24. Li, J.; Gao, Y.; Shang, S.; Rao, X.J.; Wang, Z. RSC Adv. 2014, 4, 58190–58199
25. Thanikaivelan, P.; Subramanian, V.; Rao, J.R.; Nair, B.U. Chem. Phys. Lett. 2000, 323, 59–70.
26. Li, J.; Xiao, G.; Shang, S.; Rao, X. Lett. Drug Des. Discov. 2014, 11, 59–66.
27. Karelson, M.; Lobanov, V.S.; Katritzky, A.R. Chem. Rev. 1996, 96, 1027–1043.
28. P. Sharma; A. Kumar; S. Upadhyay; V. Sahu; J. Singh; Eur. J. Med. Chem. 2009, 44, 251–259.
29. Z. Wang; J. Song; J. Chen; Z. Song; S. Shang; Z. Jiang; Z. Han. Bioorg. Med. Chem. Lett. 2008, 18,
2854–2859.
30. Kalani, K.; Yadav, D.; Khan, F.; Srivastava, S.; Suri, N. J. Mol. Model. 2012.18, 3389–3413.
31. Okoli, C.P.; Guo, Q.J.; Adewuyi, G.O. Carbohydr. Polym. 2014. 101, 40–49.
32. Sharma, D.; Narasimhan, B.; Kumar, P.; Jalbout, A. Eur. J. Med. Chem. 2009. 44, 1119–1127.
33. Sharma, P.; Kumar, A.; Upadhyay, S.; Sahu, V.; Singh, J. Eur. J. Med. Chem. 2009. 44, 251–259.
34. Schmidt, T.J.; Nour, A.M.M.; Khalid, S.A.; Kaiser, M.; Brun, R. Molecules. 2009, 14, 2062–2076.
35. Heilmann, J.; Waseschaa, M.R.; Schmidt, T.J. Bioorg. Med. Chem. 2001, 9, 2189–2194.
20

Graphic for table of contents
21